Although insulin analogs have markedly improved glycemic control in people with diabetes, the risk of glycemic excursions remains high, including death from hypoglycemia and a multitude of major negative impacts from suboptimal control of glucose levels. This proposal seeks to overcome these problems by developing an improved generation of insulin analogs, glucose-responsive insulins (GRIs), which would directly respond to blood sugar levels. These GRIs would allow insulin to be maximally active when blood sugar levels are elevated but would inactivate insulin when blood sugar levels are low. They would therefore allow the aggressive treatment of hyperglycemia while avoiding the risk of hypoglycemia, and would provide a dramatic advance in the treatment of diabetes. We recently demonstrated the potential of this approach by developing a chemically modified insulin that was made glucose-responsive by the incorporation of a phenylboronic acid for glucose sensing. In this proposal, we will synthesize a series of next-generation insulin derivatives that are designed for increased glucose responsiveness and enhanced repression under low glucose conditions, characterize their effectiveness in biochemical and cellular assays, iterate improvements to the design, and validate their potential in an animal model. Successful outcome will result in pre-therapeutic lead compounds that provide control of insulin activity and has the potential to radically improve the treatment of diabetes.